A Prognostic Model for Electrohydraulic Servovalves

Lorenzo Borello, Matteo Dalla Vedova, Giovanni Jacazio, and Massimo Sorli
Submission Type: 
Full Paper
phmc_09_081.pdf1018.44 KBSeptember 29, 2009 - 7:58am

Electrohydraulic servovalves are critical components of the hydraulic servos and their correct operation is mandatory to ensure the proper functioning of the controlled hydraulic servosystem. The servovalve operation is often monitored by measuring the position of its spool and comparing it with the servovalve electrical input. This continuous monitor can detect hardover or oscillating failures of the servovalve, but fails short of detecting other malfunctionings and it can only signal the loss of servovalve operation. A research activity was hence performed aimed at developing a prognostic algorithm able to identity the precursors of a servovalve failure, recognize the degradation pattern and estimate the remaining useful life. A model based prognosis technique was used that fuses several information obtained by comparing the actual with the expected response of the servovalve to the electrical inputs to positively recognize a progressive degradation and determine its evolution with time. The servovalves prognostic algorithm was focused to flight control systems, but it can be as well used in other application areas. In order to assess the robustness of the prognostic algorithm a simulation test environment representative of the real environment and operating conditions of a servovalve was developed. Simulations were run in which an EHSV controlled hydraulic servoactuator was subjected to time histories of commands and loads representative of those that could be encountered during actual operation, and variations of the servovalve characteristics within their normal range were simultaneously applied. At the same time, progressive degradations of the servovalve were superimposed and the ability of the prognostics algorithm to identify them was verified and a procedure to evaluate the time-to-failure was defined. The results of the research activity showed an adequate robustness of the prognostics algorithm and a confidence was hence gained in its ability to detect servovalve degradations with minimum risk of false alarms or missed failures.

Publication Year: 
Publication Volume: 
Publication Control Number: 
Submission Keywords: 
applications: aviation
Submission Topic Areas: 
Component-level PHM
Model-based methods for fault detection, diagnostics, and prognosis
Submitted by: 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed